Project Summary Deep-brain stimulation (DBS) is shown to be very effective in alleviating the motor symptoms of Parkinson's disease (PD). However, its exact mechanism is not well-understood. Clinical studies have reported conflicting results regarding the effects of DBS, with some studies suggesting that it inhibits target neurons while some others suggest that it excites those neurons. One of the significant hurdles plaguing the study of DBS is the large artifacts caused by electrical stimulation. The large artifacts saturate the neural recorder and also make it take a long time to recover to its normal working conditions. Consequently, no reliable neural feedback can be recorded during the high frequency stimulation of DBS. We have developed a new neural recorder that does not saturate even in the presence of large artifacts. The recorder has been validated in animal studies and recently in human experiments. We propose to further develop the device to study the mechanisms of DBS: in Aim 1, we will develop an artifact-resilient neural recorder and related software suite to support intraoperative monitoring during DBS. In Aim 2, we will use the proposed device to carry out an intraoperative electrophysiological recording of the subthalamic nucleus (STN) or globus pallidus internus (GPi) in PD patients. We will compare activities of the neurons when high or low stimulation frequency is used, when stimulation is delivered ipsilaterally or contralaterally, and when different temporal patterns of DBS pulses are used. Measurements of therapeutic effects in terms of tremor power will be obtained by a wireless inertial measurement unit. We will correlate tremor power, neural responses, and stimulation parameters during DBS, which can provide new insights into the mechanisms of DBS. These insights can potentially lead to a better stimulation paradigm that can enhance the efficacy of DBS.